Tag
This paper proposes Nemotron-Labs-Diffusion-Image, a masked discrete diffusion model for high-resolution text-to-image synthesis, introducing a token-editing mechanism and grouped cross-entropy objective to improve token refinement and training efficiency.
Proposes Self-Generated T2T, a training method that aligns token editing training with inference by using the model's own predictions as error sources, improving accuracy on LLaDA2.1.
Introduces Token-to-Mask (T2M) remasking to fix generation errors in masked diffusion LMs by resetting suspect tokens to mask state instead of overwriting, yielding up to +5.92 accuracy on CMATH without extra training or parameters.